Calcipotriol or calcipotriene (structure I) [CAS 112965-21-6] shows a strong activity in inhibiting undesirable proliferation of epidermal keratinocytes [F. A. C. M. Castelijins, M. J. Gerritsen, I. M. J. J. van Vlijmen-Willems, P. J. van Erp, P. C. M. van de Kerkhof; Acta Derm. Venereol. 79, 11, 1999]. The efficiency of calcipotriol and calcipotriol monohydrate (II) in the treatment of psoriasis was shown in a number of clinical trials [D. M. Ashcroft et al.; Brit. Med. J. 320, 963-67, 2000] and calcipotriol is currently used in several commercial drug formulations.

In the preparation of calcipotriol, the specific stereochemistry for the hydroxyl group at C-24 is necessary for full expression of the biological activity. Under current methodology, the required stereochemistry is introduced by one of the following methods:
(i) non-diastereoselective reduction of C-24 keto-trienes followed by the separation of diastereomeric mixtures of the C-24 hydroxyl epimers obtained via chromatography (WO 87/00834 & M. J. Calverley; Tetrahedron, 43 (20), 4609-19, 1987);
(ii) attachment of an enantiopure C-24-hydroxyl-carrying side chain to the vitamin D skeleton (M. J. Calverley, Synlett, 157-59, 1990);
(iii) selective enzymatic esterification of one of C-24 hydroxyl epimers followed by chromatographic separation (WO 03/060094).
The non-diastereoselective reduction of C-24 keto-trienes followed by chromatographic separation of the epimeric mixture (i) is the most widely practiced procedure for obtaining the desired epimer. This reduction process yields mainly the undesired C-24 epimeric alcohol (typically about 60% of the unwanted 24-R epimer) and it is difficult to separate the desired S-epimer from such a mixture by chromatography on a production scale.
The stereoselective synthesis (ii) is still an unfavourable process for scale up due to its multi step nature and cost and due to the fact that toxic intermediates are used.
The enzymatic esterification process (iii) has the disadvantage, apart from the high cost of the enzymes employed, that it introduces, depending on the selectivity of the enzyme, 1-2 additional reaction steps which adds even further costs to the process.
The stereoselective reduction of C-24 ketones directly to the desired C-24 hydroxyl epimers has for example been described for cholesterol derivatives in WO 98/24800 and by M. Ishiguro et al., J. C. S. Chem. Comm., 115-117, 1981. The stereoselective reduction of a side chain triple bond analogue of calcipotriol with unprotected triene system using S-alpine borane has been described by by M. J. Calverly et al. in Bioorg. Med. Chem. Lett., 1841-1844, 3(9), 1993.
A major technical problem of using stereoselective reduction methods for the synthesis of calcipotriol stems from the fact that the unsaturated triene system of hitherto known intermediates for the synthesis of calcipotriol are chemically labile, such as towards Lewis acidic conditions, that they are relatively easily oxidised, and that they are usually not compatible with the typical reduction reaction conditions employed. This results in reduced yields, impure products and tedious work-up procedures, especially on large-scale.
It is an object of this invention to provide an alternative process for the synthesis of calcipotriol, which may overcome one or more of the various problems and disadvantages described above.
The present invention provides a novel process to produce diastereomerically enriched C-24 hydroxyl epimers of calcipotriol derivatives using a novel synthetic pathway comprising a stereoselective reduction step. The present invention further provides novel chemically more stable intermediates where the labile triene system is protected as sulphur dioxide adduct. By producing diastereomerically enriched C-24 hydroxyl epimers of calcipotriol derivatives the yield and the efficacy of the subsequent separation of the desired C-24 S-hydroxyl epimer may be greatly improved.